Colon cancers are deeply invasive, heterogeneously differentiated tumors. Because of its invasive nature, colon cancer remains the second leading cause of cancer mortality in the United States. Numerous cellular and molecular factors have been shown to influence colon cancer progression and metastasis. There is emerging evidence both in the stem cell and in the cancer literature that physical factors in the tumor microenvironment play a critical role in regulating cell behavior in development and cancer progression. Specifically, the rigidity of a tissue has been shown to alter cell fate in breast cancer and the topographical features of the ECM have led to changes in stem cell fate. In both cases, it is felt that the cell's response to its physical surroundings is mediated by the mechanosensing, Rho-like proteins and their downstream effectors, ROCK and mDia. The effect of the ECM microtopography on colon cancer cell behavior is unknown. Thus, we propose the mechanistic hypothesis that the ECM microtopography critically regulates colon cancer de-differentiation and invasion via Rho related signaling. In our preliminary data, we show that the microtopography changes significantly as a tumor de-differentiates such that well differentiated colon cancer ECM has more "trough-like" features and a poorly differentiated tumor has more "peak-like" features. When these features are recreated in a scaffold with low protein absorptivity, less tumorigenic colon cancer cells are more motile on the "trough-like" surface while more tumorigenic colon cancer cells are more motile on the "peak-like" surface. Furthermore, these changes induced changes in Rho/ROCK signaling that were somewhat unexpected such that both less and more tumorigenic cells cultured on a "peak-like" matrix actually had higher Rho activity and more focal adhesions as revealed by talin staining. In contrast, ROCK activity was highest in colon cancer cells cultured on "trough-like" surfaces but lower on "peak-like" surfaces. Finally, when we evaluated actin polymerization in both less and more tumorigenic colon cancer cells, we found that actin polymerized more quickly in less tumorigenic cells plated on "trough-like" surfaces and faster in more tumorigenic cells plated on "peak-like" surfaces. Taken together this data suggests that the ECM microtopography is a critical regulator of colon cancer cell behavior and that mechanical signals from the ECM promote invasive characteristics in more aggressive colon cancer cells by manipulating the actin cytoskeleton. To study this we will: 1) Determine how the extracellular matrix microtopography underlying malignant colonic epithelia changes as a tumor de-differentiates and invades, 2) Demonstrate that changes in ECM microtopography enhance Rho/ROCK activity in colon cancer cells, and 3) Elucidate how changes in ECM microtopography alter elements of the actin cytoskeleton. To do this we will use novel techniques-including AFM and SEM-S-to characterize the microtopography of the ECM synthesized by cell lines and resected human colon cancers as a function of their cellular differentiation. Surfaces of defined microtopography will be generated by optical and E-beam lithography in PDMS. Cell motility, cell adhesion, focal adhesion formation, and Rho and ROCK activities will be evaluated using both malignant and non-malignant colonic epithelial cell lines on ECMs of defined microtopography using standard molecular techniques, cell rheology, and advanced imaging. The effect of changes in the ECM microtopography on invasion will be assessed using type I collagen scaffolds. Alterations in the actin cytoskeleton will be characterized using a series of actin mutants in both live cell assays and standard actin assays. Overall, these studies will allow us to understand the contribution of the extracellular matrix topography to post neoplastic transformation and invasion at the level of individual cells in colon cancer. This project is dedicated to elucidating the effect of changes in the extracellular matrix microtopography on colon cancer de-differentiation and invasion. This project is relevant to human health and disease insofar as colon cancer is the second leading cause of colon cancer death in the United States. This is because of our limited ability to treat metastatic disease. Thus, this project will provide a new approach to the study of colon cancer invasion and in doing so may help us better identify patients at risk for metastasis as well as those in whom certain chemotherapeutic regimens would create a tumor microenvironment more conducive to metastasis.